DKDavid A. Kenny

David A. Kenny

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


David A. Kenny has a total of 113 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    The Social-Relations Model for Asymmetric-Block-Design Data: A Tutorial With R

    by Avraham N. Kluger, Robert A. Ackerman, David A. Kenny, Thomas E. Malloy, Paul W. Eastwick

    We provide a guide to estimating social-relations-model (SRM) parameters for data collected with the asymmetric block design using R. SRM estimates reflect how people differ in the social behaviors they emit and elicit from others and the extent to which they behave differently with each unique partner. Such analyses have proven useful in studies of the power of leaders, negotiation effectiveness, and personality perception, to name a few. The SRM parameters can be estimated for data collected with various designs; the most common design is the round-robin. A less used design in SRM studies is the asymmetric block design, which involves measurements taken from dyadic interactions of individuals from two different social categories (e.g., speed-dating studies in which different-gendered participants interact). In this guide, we show how to estimate SRM parameters for asymmetric block design by specifying mixed-effect models with two R packages: glmmTMB and brms. Specifically, we show how to calculate SRM estimates for (a) one continuous outcome, (b) multiple measures of a continuous outcome, (c) bivariate correlations for two continuous outcomes, (d) one dichotomous outcome, (e) bivariate and multivariate associations between continuous SRM variables and one dichotomous outcome (in supplementary work-around), and (f) data from mother and child interactions rather than the more common speed-dating data. In all analyses, we illustrate how to model both fixed and random effects to allow testing additional fixed effects, such as group characteristics and the order of interaction between participants.


  • Parent–Child Relationships From Adolescence to Early Adulthood: The Role of Conceptions of Adulthood

    by W. S. Aquilino, J. J. Arnett, J. L. Tanner, L. M. Padilla-Walker, S. Badger, L. J. Nelson, C. M. Barry, P. B. Baltes, R. M. Baron, D. A. Kenny, et al.

    Since the 1990s, terms such as ‘parasite singles’, ‘the kangaroo generation’ and ‘NEET’ (Not in Employment, Education or Training) have been used by the mass media of Japan, South Korea and China to address the topic of whether young adults of this generation are refusing to grow up. This topic has aroused heated discussion. Japanese scholar Yamada coins the term ‘parasite singles’ to refer to the many unmarried adults in Japan who continue to live with their parents. These young adults resemble parasites by living off their parents with low ...

  • Enhancing validity in psychological research.

    by David A. Kenny

    Methods to increase Campbell's (1957) internal and external validity as well as Cook and Campbell's (1979) construct and conclusion validity are reviewed. For internal validity or valid causal inference, designs and methods to draw causal conclusions from nonrandomized studies are considered. Greater collaboration between the causal inference and structural equation modeling traditions would benefit both. For external validity, generalizing results, treating partners and studies as well as participants as random is strongly encouraged. For construct validity, particularly the psychological meaning of measures, multivariate models that treat measures from both overtime and dyadic data as being a combination of multiple constructs are discussed. For conclusion validity or valid statistical inference, the problem of low power when generalizability is high and the assumption of independence are discussed. Finding the truth in psychological research is a challenge, and seemingly insurmountable difficulties are often encountered. Nonetheless, persistent and diligent efforts using strategies developed by generations of methodologists should lead to scientific advancement. (PsycINFO Database Record (c) 2019 APA, all rights reserved).


  • Reflections on the actor–partner interdependence model

    by David A. Kenny

    The actor–partner interdependence model (APIM) has been widely successful and has become the default method for analyses of dyadic data in which both members of the pair have the same measurements. This article reviews the history of the development of the APIM, focusing on the original presentation of the model, a 1979 article by Kraemer and Jacklin. Also discussed are the topics of APIM patterns, whether the APIM is a dyadic model, and the automation of APIM analyses. Although the APIM is the most popular method for dyadic analysis, it should be recognized there are alternatives that should also be considered.


  • Top Cited Papers

    Experiments with More Than One Random Factor: Designs, Analytic Models, and Statistical Power

    by Charles M. Judd, Jacob Westfall, David A. Kenny

    Traditional methods of analyzing data from psychological experiments are based on the assumption that there is a single random factor (normally participants) to which generalization is sought. However, many studies involve at least two random factors (e.g., participants and the targets to which they respond, such as words, pictures, or individuals). The application of traditional analytic methods to the data from such studies can result in serious bias in testing experimental effects. In this review, we develop a comprehensive typology of designs involving two random factors, which may be either crossed or nested, and one fixed factor, condition. We present appropriate linear mixed models for all designs and develop effect size measures. We provide the tools for power estimation for all designs. We then discuss issues of design choice, highlighting power and feasibility considerations. Our goal is to encourage appropriate analytic methods that produce replicable results for studies involving new samples of both participants and targets.


  • A toolbox with programs to restructure and describe dyadic data

    by Thomas Ledermann, David A. Kenny

    When one gathers dyadic data, one is very often faced with the burdensome task of restructuring the data. For instance, the use of multiple regression analysis or structural equation modeling (SEM) requires one type of data structure, whereas multilevel modeling or multilevel SEM usually requires a different data structure. However, data are often entered in neither of these structures. In this article, we first describe the most typical dyadic data formats, what format the major data-analytic methods require, and then present a toolbox called restructure and describe dyadic data (RDDD) with programs that restructure dyadic data from one format into another. Moreover, the programs identify different types of dyadic variables and provide descriptive and inferential statistics that can be informative to dyadic researchers. The programs, written in R, provide a graphical user interface and are designed to work with minimal input information that is much less than standard restructuring procedures.


  • Top Cited Papers

    The Performance of RMSEA in Models With Small Degrees of Freedom

    by David A. Kenny, Burcu Kaniskan, D. Betsy McCoach

    Given that the root mean square error of approximation (RMSEA) is currently one of the most popular measures of goodness-of-model fit within structural equation modeling (SEM), it is important to know how well the RMSEA performs in models with small degrees of freedom (df). Unfortunately, most previous work on the RMSEA and its confidence interval has focused on models with a large df. Building on the work of Chen et al. to examine the impact of small df on the RMSEA, we conducted a theoretical analysis and a Monte Carlo simulation using correctly specified models with varying df and sample size. The results of our investigation indicate that when the cutoff values are used to assess the fit of the properly specified models with small df and small sample size, the RMSEA too often falsely indicates a poor fitting model. We recommend not computing the RMSEA for small df models, especially those with small sample sizes, but rather estimating parameters that were not originally specified in the model.


  • Power Anomalies in Testing Mediation

    by David A. Kenny, Charles M. Judd

    Two rather surprising anomalies relating to statistical power occur in testing mediation. First, in a model with no direct effect for which the total effect and indirect effect are identical, the power for the test of the total effect can be dramatically smaller than the power for the test of the indirect effect. Second, when there is a direct effect of a causal variable on the outcome controlling for the mediator, the power of the test of the indirect effect is often considerably greater than the power of the test of the direct effect, even when the two are of the same magnitude. We try to explain the reasons for these anomalies and how they affect practice.